转录组基因分类

yy951010

import pandas as pd
# 加载数据
data = pd.read_csv("C:/Users/Administrator/Desktop/proteinfpkmshaixuan.txt", sep="\t")  # 替换为实际文件路径
# 提取基因ID
gene_ids = data['id']
data = data.set_index('id')
# 1. NotExpressed（未表达）：所有组织中表达水平低于 1 的基因
not_expressed = data[(data < 1).all(axis=1)]
# 2. Tissue Enriched（组织富集）：某组织中表达量 ≥1，且比其他所有组织高至少 5 倍
tissue_enriched = []
for tissue in data.columns:
    enriched = data[
        (data[tissue] >= 1) &
        (data[tissue] >= 5 * data.drop(columns=[tissue]).max(axis=1))
    ]
    enriched['tissue'] = tissue
    tissue_enriched.append(enriched)
tissue_enriched = pd.concat(tissue_enriched) if tissue_enriched else pd.DataFrame()
# 3. Group Enriched（集团丰富）：一组 2-7 个组织中表达量 ≥1，且比其他所有组织高至少 5 倍
from itertools import combinations
group_enriched = []
for tissue_count in range(2, 8):
    for tissues in combinations(data.columns, tissue_count):
        other_tissues = data.columns.difference(tissues)
        enriched = data[
            (data[list(tissues)].min(axis=1) >= 1) &
            (data[list(tissues)].min(axis=1) >= 5 * data[other_tissues].max(axis=1))
        ]
        enriched['group'] = ', '.join(tissues)
        group_enriched.append(enriched)
group_enriched = pd.concat(group_enriched) if group_enriched else pd.DataFrame()
# 4. Tissue Enhanced（组织增强）：某组织中表达量 ≥1，且比其他组织平均值高至少 5 倍
tissue_enhanced = []
for tissue in data.columns:
    enhanced = data[
        (data[tissue] >= 1) &
        (data[tissue] >= 5 * data.drop(columns=[tissue]).mean(axis=1))
    ]
    enhanced = enhanced[
        ~enhanced.index.isin(tissue_enriched.index) &
        ~enhanced.index.isin(group_enriched.index)
    ]
    enhanced['tissue'] = tissue
    tissue_enhanced.append(enhanced)
tissue_enhanced = pd.concat(tissue_enhanced) if tissue_enhanced else pd.DataFrame()
# 5. Expressed in all（全部表达）：在所有组织中表达量 ≥1 且不属于其他分类
expressed_in_all = data[
    (data >= 1).all(axis=1) &
    (~data.index.isin(not_expressed.index)) &
    (~data.index.isin(tissue_enriched.index)) &
    (~data.index.isin(group_enriched.index)) &
    (~data.index.isin(tissue_enhanced.index))
]
# 6. Mixed（混合）：不属于上述任何一类
mixed = data[
    ~data.index.isin(pd.concat([not_expressed, tissue_enriched, group_enriched, tissue_enhanced, expressed_in_all]).index)
]
# 输出分类结果
not_expressed.to_csv("NotExpressed_genes.csv")
tissue_enriched.to_csv("TissueEnriched_genes.csv")
group_enriched.to_csv("GroupEnriched_genes.csv")
tissue_enhanced.to_csv("TissueEnhanced_genes.csv")
expressed_in_all.to_csv("ExpressedInAll_genes.csv")
mixed.to_csv("Mixed_genes.csv")
print("分类完成，结果已保存到文件。")

田田爱崽崽

感谢大佬分享哦

sx4267

谢谢分享

yy951010

sx4267 发表于 2025-1-8 11:53
谢谢分享

不客气，互相学习

yy951010

田田爱崽崽 发表于 2025-1-8 08:29
感谢大佬分享哦

不客气，互相学习哦

wooailein

这个专业性就高一截了